JP5013436B2 - Power semiconductor device - Google Patents

Abstract

A semiconductor layer has a first layer of first conductive type, a second layer of second conductive type, and a third layer. The third layer has a first region of first conductive type, and a second region of second conductive type. A second electrode is in contact with each of the first and second regions. A trench is formed on the semiconductor layer at a surface opposite to its surface facing a first electrode. A gate electrode is embedded in the trench with a gate insulating film interposed therebetween. The gate electrode includes a first portion projecting into the first layer through the first region and the second layer, a second portion projecting into the first layer through the second region and the second layer. The second portion projects into the first layer deeper than a depth in which the first portion projects into the first layer.

Description

  The present invention relates to a power semiconductor device, and more particularly to a power semiconductor device having a gate electrode embedded in a trench.

  Some power semiconductor devices are used as contactless switches for controlling large-capacity power. Such a large-capacity device is applied to, for example, an inverter circuit of home appliances such as an air conditioner, a refrigerator, and a washing machine, which are saving energy, and is applied to a motor control of a train such as a bullet train or a subway. In recent years, considering the global environment, power semiconductor devices have been used for inverter / converter control in hybrid cars that run on electric motors and engines, and for converters for photovoltaic or wind power generation. Have been. Thus, the application fields of power semiconductor devices are expanding.

  An example of such a power semiconductor device is an IGBT (Insulated Gate Bipolar Transistor). The IGBT is a typical switching element that controls a large current with low loss.

Here, the operation principle of the IGBT will be briefly described.
First, turn-on will be described. When a sufficient positive voltage (for example, +15 V) is applied between the gate and the emitter, a MOSFET (Metal Oxide Semiconductor Field Effect Transistor) located on the surface side of the IGBT is turned on. Then, a forward bias is applied between the collector p ⁺ layer and the n ⁻ drift layer, and holes are injected from the p ⁺ layer to the n ⁻ layer. Then the amount of electrons corresponding to the charge amount of injected holes the n ^- concentrated in the drift layer, n ^- resistance reduction of the drift layer (conductivity modulation) occurs. As a result, the IGBT is turned on.

Second, turn-off will be described. When the voltage between the gate and the emitter is lowered, the MOSFET located on the surface side of the IGBT is turned off. Then, the hole injection from the collector p ⁺ layer stops and the n ⁻ drift layer is depleted, so that the holes that have already been injected flow out to the emitter side and the current is cut off.

Here, the decrease in resistance of the n ⁻ drift layer due to the conductivity modulation in the ON state means a reduction in the resistance of the device, and the voltage between the collector and the emitter at that time is called “ON voltage”. In addition, a current due to residual holes at the time of turn-off becomes a switching loss. That is, as holes and electrons (collectively referred to as carriers) are injected into the n ⁻ drift layer in order to reduce the resistance, loss due to carrier remaining (switching loss) increases at the time of turn-off. That is, there is a trade-off relationship between the ON voltage and the switching loss.

  In order to improve this trade-off characteristic, a trench IGBT in which the transistor cell density is improved by using a miniaturization technique is disclosed. A trench IGBT has a gate electrode embedded in a trench formed on a semiconductor layer via a gate insulating film. A technique for forming a trench is disclosed in, for example, Japanese Patent Laid-Open No. 6-291178 (Patent Document 1). In addition to IGBTs, CSTBTs (Carrier Stored Trench-gate Bipolar Transistors) and IEGTs (Injection Enhanced Gate Transistors) with improved carrier density in the drift layer have been developed.

By the way, when an unexpected operation such as a load short circuit or an arm short circuit occurs, a large current / high voltage is applied to the IGBT. Even in such a case, the IGBT element needs to be able to withstand a certain amount of energy. In the process where the gate is turned off when the short circuit occurs, the collector voltage rises, and the current decays, the carriers (holes) accumulated in the n ⁻ drift layer correspond to dv / dt, that is, the time differential value of the collector-emitter voltage. Then discharged. When the hole current path at that time flows through the base region of the npn parasitic transistor of the MOSFET, there is a problem that the IGBT is easily latched up.

  As a technique for preventing latch-up, for example, there is one disclosed in Japanese Patent Application Laid-Open No. 2008-21918 (Patent Document 2). According to this publication, a semiconductor device includes a first conductivity type collector layer, a second conductivity type semiconductor layer, a first conductivity type base region, a second conductivity type emitter region, a first trench, , A first gate electrode, a second trench, a second gate electrode, an emitter electrode connected to the base region and the emitter region, and a collector electrode connected to the collector layer. The semiconductor layer is formed on the collector layer. The base region is formed on the surface of the semiconductor layer. The emitter region is formed on a part of the surface of the base region. The first trench is dug until reaching the semiconductor layer from the surface of the emitter region. The first gate electrode is embedded in the first trench through the first insulating film. The second trench is dug until reaching the semiconductor layer from the surface of the base region other than the emitter region. The second gate electrode is embedded in the second trench via a second insulating film. The second trench is deeper than the first trench.

JP-A-6-291178 JP 2008-21918 A

  According to the technique disclosed in Japanese Patent Application Laid-Open No. 2008-21918, in addition to the first gate electrode that is the original gate electrode of the IGBT, a second gate, that is, a dedicated gate for preventing latch-up must be provided separately. I must. As a result, since the structure of the IGBT is greatly changed, there is a problem that the electrical characteristics of the IGBT are also greatly changed.

  The present invention has been made in view of the above problems, and an object thereof is to provide a power semiconductor device capable of preventing the occurrence of latch-up.

  The power semiconductor device of the present invention includes first and second electrodes, a semiconductor layer, a gate insulating film, and a gate electrode. The semiconductor layer is provided on the first electrode. The semiconductor layer includes a first conductivity type first layer, a second conductivity type second layer, and a third layer. The first layer is provided on the first electrode. The second layer is provided on the first layer. The third layer is provided on the second layer. The third layer has a first conductivity type first region and a second conductivity type second region. The second electrode is in contact with each of the first and second regions. A trench is formed on the surface of the semiconductor layer opposite to the surface facing the first electrode. The gate insulating film covers the inner wall of the trench. The gate electrode is embedded in the trench through a gate insulating film. The gate electrode penetrates the first region through the first region and the second layer, and penetrates the first layer through the second region and the second layer. A second portion. The second portion penetrates deeply into the first layer as compared to the depth at which the first portion penetrates into the first layer.

  According to the power semiconductor device of the present invention, the second portion of the gate electrode penetrates deeper into the first layer than the depth at which the first portion of the gate electrode penetrates into the first layer. ing. As a result, the current flowing in the vicinity of the second portion increases, and conversely, the current flowing in the vicinity of the first portion decreases. Therefore, the current flowing in the in-plane direction in the second layer immediately below the first region can be reduced. Thereby, occurrence of latch-up can be prevented.

1 is a partial plan view schematically showing a configuration of a power semiconductor device according to a first embodiment of the present invention. FIG. 2 is a schematic partial sectional view taken along line II-II in FIG. 1. FIG. 3 is a schematic partial sectional view taken along line III-III in FIG. 1. FIG. 4 is a schematic partial cross-sectional view taken along line IV-IV in FIG. 1. FIG. 5 is a schematic partial cross-sectional view taken along line VV in FIG. 1. FIG. 4 is a diagram schematically showing an impurity concentration profile along an arrow VI in FIG. 3. FIG. 5 is a diagram schematically showing an impurity concentration profile along arrow VII in FIG. 4. FIG. 4 is a partially enlarged view of FIG. 3 and shows the behavior of hole current at turn-off. FIG. 5 is a partially enlarged view of FIG. 4 and shows the behavior of the hole current at turn-off. It is a graph which shows the simulation result of the relationship between the depth of the trench of IGBT, and the intensity profile of an electric field. It is a partial top view which shows roughly the structure of the semiconductor device for electric power of a comparative example. FIG. 12 is a schematic partial cross-sectional view taken along line XII-XII in FIG. 11. FIG. 12 is a schematic partial cross-sectional view taken along line XIII-XIII in FIG. 11. FIG. 14 is a schematic partial cross-sectional view taken along line XIV-XIV in FIG. 11. FIG. 14 is a partially enlarged view of FIG. 13 and shows the behavior of the hole current at turn-off. FIG. 15 is a partially enlarged view of FIG. 14 showing the behavior of hole current at turn-off. It is a fragmentary top view which shows roughly the structure of the semiconductor device for electric power in Embodiment 2 of this invention. FIG. 18 is a schematic partial cross-sectional view taken along line XVIII-XVIII in FIG. 17. FIG. 18 is a schematic partial cross-sectional view taken along line XIX-XIX in FIG. 17. FIG. 18 is a schematic partial cross-sectional view taken along line XX-XX in FIG. 17. FIG. 20 is a diagram schematically showing an impurity concentration profile along arrow XXI in FIG. 19. FIG. 21 is a diagram schematically showing an impurity concentration profile along arrow XXII in FIG. 20. FIG. 20 is a partially enlarged view of FIG. 19 showing the behavior of the hole current at turn-off. FIG. 21 is a partially enlarged view of FIG. 20, showing the behavior of hole current at turn-off. It is a fragmentary top view which shows schematically the structure of the power semiconductor device in Embodiment 3 of this invention. FIG. 26 is a schematic partial sectional view taken along line XXVI-XXVI in FIG. 25. FIG. 26 is a schematic partial cross-sectional view taken along line XXVII-XXVII in FIG. 25. FIG. 26 is a schematic partial sectional view taken along a line XXVIII-XXVIII in FIG. 25. It is a fragmentary top view which shows roughly the structure of the power semiconductor device in Embodiment 4 of this invention. FIG. 30 is a schematic partial cross-sectional view taken along line XXX-XXX in FIG. 29. FIG. 30 is a schematic partial cross-sectional view taken along line XXXI-XXXI in FIG. 29. FIG. 30 is a schematic partial cross-sectional view taken along line XXXII-XXXII in FIG. 29. FIG. 32 schematically shows an impurity concentration profile along arrow XXXIII in FIG. 31. FIG. 33 is a diagram schematically showing an impurity concentration profile along arrow XXXIV in FIG. 32. FIG. 10 is a partial plan view schematically showing a configuration of a power semiconductor device in a fifth embodiment of the present invention. FIG. 36 is a schematic partial cross-sectional view taken along line XXXVI-XXXVI in FIG. 35. FIG. 36 is a schematic partial cross-sectional view taken along line XXXVII-XXXVII in FIG. 35. FIG. 36 is a schematic partial cross-sectional view taken along line XXXVIII-XXXVIII in FIG. 35. FIG. 38 is a diagram schematically showing an impurity concentration profile along arrow XXXIX in FIG. 37. FIG. 39 is a diagram schematically showing an impurity concentration profile along arrow XL in FIG. 38. FIG. 38 is a partially enlarged view of FIG. 37, showing the behavior of hole current at turn-off. FIG. 39 is a partially enlarged view of FIG. 38, illustrating the behavior of hole current at turn-off. It is a fragmentary top view which shows roughly the structure of the power semiconductor device in Embodiment 6 of this invention. FIG. 44 is a schematic partial cross-sectional view taken along line XLIV-XLIV in FIG. 43. FIG. 44 is a schematic partial cross-sectional view taken along line XLV-XLV in FIG. 43. FIG. 44 is a schematic partial cross-sectional view taken along line XLVI-XLVI in FIG. 43. It is a fragmentary top view which shows schematically the structure of the semiconductor device for electric power in Embodiment 7 of this invention. FIG. 48 is a schematic partial cross-sectional view taken along line XLVIII-XLVIII in FIG. 47. FIG. 48 is a schematic partial cross-sectional view taken along line XLIX-XLIX in FIG. 47. FIG. 48 is a schematic partial cross-sectional view taken along line LL in FIG. 47. FIG. 50 is a diagram schematically showing an impurity concentration profile along an arrow LI in FIG. 49. The impurity concentration profile along the arrow LII in FIG. 50 is a diagram schematically showing. It is a fragmentary sectional view which shows schematically the structure of the power semiconductor device in Embodiment 8 of this invention. It is a fragmentary top view which shows schematically the structure of the semiconductor device for electric power in Embodiment 9 of this invention. FIG. 55 is a schematic partial cross-sectional view taken along line LV-LV in FIG. 54. FIG. 55 is a schematic partial cross-sectional view taken along line LVI-LVI in FIG. 54. FIG. 57 is a schematic partial cross-sectional view taken along line LVII-LVII in FIG. 54. FIG. 57 is a schematic partial cross-sectional view taken along line LVIII-LVIII in FIG. 54. FIG. 58 is a diagram schematically showing an impurity concentration profile along arrow LIX in FIG. 57. FIG. 59 is a diagram schematically showing an impurity concentration profile along an arrow LX in FIG. 58. It is a fragmentary top view which shows schematically the structure of the semiconductor device for electric power in Embodiment 10 of this invention. FIG. 62 is a schematic partial cross-sectional view taken along line LXII-LXII in FIG. 61. FIG. 62 is a schematic partial cross-sectional view taken along line LXIII-LXIII in FIG. 61. FIG. 62 is a schematic partial cross-sectional view taken along line LXIV-LXIV in FIG. 61. FIG. 62 is a schematic partial cross-sectional view taken along line LXV-LXV in FIG. 61. FIG. 65 is a diagram schematically showing an impurity concentration profile along arrow LXVI in FIG. 64. FIG. 66 schematically shows an impurity concentration profile taken along arrow LXVII in FIG. 65. It is a fragmentary top view which shows schematically the structure of the power semiconductor device in Embodiment 11 of this invention. FIG. 69 is a schematic partial cross-sectional view taken along line LXIX-LXIX of FIG. 68. FIG. 69 is a schematic partial cross-sectional view taken along line LXX-LXX in FIG. 68. FIG. 69 is a schematic partial cross-sectional view taken along line LXXI-LXXI of FIG. 68. FIG. 69 is a schematic partial cross-sectional view taken along line LXXII-LXXII of FIG. 68. FIG. 72 schematically shows an impurity concentration profile along arrow LXXIII in FIG. 71. FIG. 73 is a diagram schematically showing an impurity concentration profile along an arrow LXXIV in FIG. 72.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
In some drawings, the coordinate axes of the XYZ coordinate system are shown. The X direction is the extending direction of the gate electrode, the Y direction is the extending direction of each stripe in the stripe cell of the transistor, and the Z direction is the thickness direction. The zero point of the Z axis is the position of the interface between the third layer (n ⁺ source region and p ⁺ contact region) and the emitter electrode 11, and the positive direction of the Z axis is the direction from the zero point toward the semiconductor layer. It is.

(Embodiment 1)
With reference to FIGS. 1-5, the structure of IGBT101 as a power semiconductor device of this Embodiment is demonstrated. FIG. 1 is a diagram showing a transistor cell of the IGBT 101 from the emitter side. In FIG. 1, an emitter electrode 11, an interlayer insulating film 10, and a gate insulating film 9 to be described later are shown for easy understanding of the drawing. Absent.

The IGBT 101 includes a collector electrode 12 (first electrode), an emitter electrode 11 (second electrode), a gate insulating film 9, a gate electrode EV, an interlayer insulating film 10, and a semiconductor layer. The semiconductor layer is provided on the collector electrode 12, and includes a p-type (second conductivity type) p ⁺ collector layer 6 (fourth layer) and an n-type (first conductivity type) n ⁺ buffer layer 7. An n-type n ⁻ drift layer 8 (first layer), a p-type (second conductivity type) p base layer 14 (second layer), and a third layer described below. .

The third layer has an n-type n ⁺ source region 2 (first region) and a p-type p ⁺ contact region 3 (second region). N ⁺ source region 2 and p ⁺ contact region 3 are formed in stripes in a direction perpendicular to trench 5V in plan view (FIG. 1). That is, the n ⁺ source region 2 and the p ⁺ contact region 3 constitute a so-called stripe cell.

The p ⁺ collector layer 6 is provided on the collector electrode 12. The n ⁺ buffer layer 7 is provided on the p ⁺ collector layer 6. The n ⁻ drift layer 8 is provided on the n ⁺ buffer layer 7. That is, n ⁻ drift layer 8 is provided on collector electrode 12 via p ⁺ collector layer 6 and n ⁺ buffer layer 7. The p base layer 14 is provided on the n ⁻ drift layer 8. The third layer, that is, the n ⁺ source region 2 and the p ⁺ contact region 3 are provided on the p base layer 14.

  A trench 5V is formed on the surface opposite to the surface facing the collector electrode 12 of the semiconductor layer (the surface shown in FIG. 1). The gate insulating film 9 covers the inner wall of the trench 5V. The gate electrode EV is embedded in the trench 5V through the gate insulating film 9.

Gate electrode EV penetrates n ⁺ source region 2 and p base layer 14 to enter n ⁻ drift layer 8, and p ⁺ contact region 3 and p base layer 14 penetrate n ⁻ And a second portion 13 penetrating into the drift layer 8. Moreover, the 1st part 1 and the 2nd part 13 are integrally formed. That is, the gate electrode EV is provided so as to cross the stripe-like arrangement of the n ⁺ source region 2 and the p ⁺ contact region 3 in plan view. Thus, the n ⁺ source region 2 and the p ⁺ contact region 3 are configured to have the same potential.

The first portion 1 and the second portion 13 of the gate electrode EV have thicknesses D1 and D2 (FIG. 2), respectively. The thickness D2 is larger than the thickness D1. Thus the first portion 1 (FIG. 3) the n ^- than the depth to penetrate the drift layer 8, a second portion 13 (FIG. 4) is the n ^- penetrates deep into the drift layer 8. A portion of trench 5V in which each of first portion 1 and second portion 13 is embedded has widths W1 and W2 (FIG. 1). Accordingly, the portion of the trench 5V in which the second portion 13 is embedded is wider than the portion of the trench 5V in which the first portion 1 is embedded.

Emitter electrode 11 is in contact with each of n ⁺ source region 2 and p ⁺ contact region 3 at the position of emitter contact 4 (FIGS. 3 and 4). The emitter electrode 11 is insulated from the gate electrode EV by the interlayer insulating film 10.

As specific dimensions, for a 600V class IGBT, for example, the thickness D1 is about 6 μm and the thickness D2 is about 7 μm. That is, the depth of the portion of the trench 5V in which the first portion 1 of the gate electrode EV is embedded is about 6 μm, and the depth of the portion of the trench 5V in which the second portion 13 of the gate electrode EV is embedded is about 7 μm. The width W1 of the trench 5V is about 1 μm and the width W2 is about 1.4 μm.
Each of the n ⁺ source region 2 and the p ⁺ contact region 3 has a thickness of about 1 μm, and the p base layer 14 has a thickness of about 3 μm.

Referring to FIGS. 3 and 6 , when the peak concentration in the impurity concentration profile along arrow V I (FIG. 3) is exemplified as the number of ions per unit volume, the peak concentration of n ⁺ source region 2 is 1 × 10 ¹⁹ / cm ³ , the p base layer 14 has a peak concentration of 5 × 10 ¹⁷ / cm ³ , the n ⁻ drift layer 8 has a peak concentration of 1.5 × 10 ¹⁴ / cm ³ , and the n ⁺ buffer layer 7 has a peak concentration of 1 × 10 5. The peak concentration of ¹⁶ / cm ³ and p ⁺ collector layer 6 is 1 × 10 ¹⁹ cm ³ .

Referring to FIGS. 4 and 7 , when the peak concentration in the impurity concentration profile along arrow VI I (FIG. 4) is exemplified as the number of ions per unit volume, the peak concentration of p ⁺ contact region 3 is 1 × 10 ¹⁹ / cm ³ , the p base layer 14 has a peak concentration of 5 × 10 ¹⁷ / cm ³ , the n ⁻ drift layer 8 has a peak concentration of 1.5 × 10 ¹⁴ / cm ³ , and the n ⁺ buffer layer 7 has a peak concentration of 1 × 10 5. The peak concentration of ¹⁶ / cm ³ and p ⁺ collector layer 6 is 1 × 10 ¹⁹ / cm ³ .

Next, an outline of a method for forming the gate electrode EV will be described.
Referring to FIG. 1, first, a semiconductor layer is prepared, and an etching mask having an opening corresponding to the planar pattern of trench 5V is formed on the semiconductor layer. Next, using this etching mask, the semiconductor layer is etched by, for example, a dry etching method. In this etching, due to the microloading effect, the portion where the width of the opening is wider is etched deeper than the portion where the opening is narrow. That is, the width W2 portion of the trench 5V is etched deeper than the width W1 portion of the trench 5V. Specifically, in the trench 5V, for example, the depth of the portion having the width W1 = 1 μm is 6 μm, the depth of the portion having the width W2 = 1.4 μm is 7 μm, and the difference in depth of 1 μm is caused by the microloading effect. appear.

  Next, gate insulating film 9 is formed so as to cover the inner wall of trench 5V. Then, the gate electrode EV is formed so as to fill the trench 5V via the gate insulating film 9.

  As described above, as the gate electrode EV, the first portion 1 having the thickness D1 is formed in the width W1 portion of the trench 5V, and the second portion D2 having the thickness D2 in the width W2 portion of the trench 5V. A portion 13 is formed.

Next, the operation of the IGBT 101 will be described.
First, the on state of the IGBT 101 will be described. In the on state, a current flows from the collector electrode 12 side to the emitter electrode 11 side as shown by an arrow 15 (FIG. 3).

Next, the behavior of the hole current when the IGBT 101 is turned off will be described. Ideally, the hole current at the time of turn-off escapes to the p ⁺ contact region 3 along the Z direction as shown by an arrow 20d (FIG. 4). In practice, however, the hole current flows in a more complex manner, as will be explained below.

Referring mainly to FIG. 8, at the time of turn-off, holes filling n ⁻ drift layer 8 are discharged to emitter electrode 11 (FIG. 3) side by a depletion layer extending from p base layer 14. At this time, due to the strong electric field region 16 generated at the bottom of the first portion 1 of the gate electrode EV (FIG. 2), some hole currents of the first portion 1 It passes through the side wall of the first part 1 via the bottom.

  Referring to FIG. 9, another part of the hole current is generated by a strong electric field region 19 generated at the bottom of the second portion 13 of the gate electrode EV (FIG. 2) as indicated by an arrow 20 in the drawing. , Flows through the bottom of the second portion 13, passes through the side wall of the second portion 13, and flows to the emitter side.

  When the current density values corresponding to each of the arrow 17 (FIG. 8) and the arrow 20 (FIG. 9) are compared, the current density corresponding to the arrow 17 is smaller. This is because the first portion 1 of the gate electrode EV (FIG. 2) is shallower than the second portion 13, so that the electric field in the strong electric field region 16 is more than the electric field in the strong electric field region 19. This is because it becomes smaller. Hereinafter, the verification result of the magnitude relation of the electric field will be described.

Referring to FIG. 10, the positions along the Z direction (thickness direction) (broken arrows in FIG. 9) in each case where the trench depth is 1.6 μm, 1.8 μm, 2.0 μm, and 2.2 μm. The result of simulating the electric field intensity E in the Z direction at the position along () is shown as intensity profiles G1 to G4. As a result, both the electric field strength RI in the gate insulating film 9 and the electric field strength RS in the n ⁻ drift layer 8 at the bottom of the trench are smaller when the trench is shallower.

  As described above, since the electric field in the strong electric field region 16 (FIG. 8) is smaller than the electric field in the strong electric field region 19 (FIG. 9), the hole current at the turn-off time is the arrow 17 (FIG. 8) and the arrow 20 Of the routes shown in FIG. 9, the route shown mainly by the arrow 20 is passed. That is, the hole current passing through the path indicated by the arrow 17 becomes small due to the difference in electric field strength.

Further, referring to FIG. 5, the hole current that has passed through the path indicated by arrow 17 (FIG. 8) is a silicon mesa region, that is, a pair of trenches 5V (FIG. 1) as indicated by arrow 17m. Flows through an area extending over n ⁺ source length SL. This current flows to the emitter side in the parasitic npn transistor 120 via the base region having the p base length BL. That is, this current contributes to the base current of the parasitic npn transistor 120.

If this base current becomes excessive, a latch-up current 121 flows from the collector side to the emitter side in the parasitic npn transistor 120. That is, latch-up occurs. In particular, when the pinch resistor 18 (Rpin) is large, the amplification factor h _FE of the parasitic npn transistor 120 may increase, and latch-up breakdown may occur.

  However, according to the present embodiment, since the hole current passing through the path indicated by the arrow 17 (FIG. 8) is small as described above, the current flowing following the arrow 17, that is, the hole passing through the path indicated by the arrow 17m (FIG. 5). The current is small. As a result, since the base current flowing through the parasitic npn transistor 120 is suppressed, the latch-up of the IGBT 101 is suppressed.

  The hole current indicated by the arrow 20 flows through the region where the parasitic npn transistor 120 is not formed and is discharged to the emitter side, and therefore does not cause latch-up.

Next, the IGBT 100 of the comparative example will be described.
Referring to FIG. 11, a trench 5S is formed in the semiconductor layer of IGBT 100 of the comparative example. The entire trench 5S has a width W1. That is, the trench 5S has a uniform width.

12 to 14, the gate electrode ES is embedded in the trench 5 </ b> S (FIG. 11) via the gate insulating film 9. Entire gate electrode ES has a thickness D1 (Fig. 1 2). That is, the gate electrode ES has a uniform thickness.

  Referring to FIGS. 15 and 16, due to the above thickness uniformity, strong electric field regions 16Z (FIG. 15) and 19Z (FIG. 16) are substantially the same. Therefore, at the time of turn-off, a large hole current flows not only through the path indicated by the arrow 20Z (FIG. 16) but also through the path indicated by the arrow 17Z (FIG. 15). As a result, the base current of the parasitic npn transistor 120 (FIG. 5) becomes excessive, so that the IGBT 100 is easily latched up.

Particularly when the turn-off time is fast, that is, when the time derivative (dV / dt) of the collector-emitter voltage is large, the hole current is not smoothly discharged to the emitter side, and holes can be accumulated in the p base layer 14. Therefore, the IGBT 100 is likely to be destroyed. Even when the collector current is large, the carrier current accumulated in the n ⁻ drift layer 8 is large, so that the hole current at the time of turn-off becomes large, so that the IGBT 100 is likely to be destroyed.

If the purpose is only to suppress the occurrence of latch-up, the pinch resistor 18 (FIG. 15) of the IGBT 100 may be reduced. To this end, for example, three countermeasures are firstly reducing the n ⁺ source length SL, secondly increasing the p base length BL, or thirdly increasing the impurity concentration of the p base layer 14. Can be considered. The first countermeasure requires a more advanced microfabrication technique, and when the n ⁺ source length SL is excessively reduced, the variation in threshold voltage and on-voltage increases. The second countermeasure is to increase the on-voltage by increasing the channel resistance. The third countermeasure increases the threshold voltage. Thus, if it is attempted to suppress the occurrence of latch-up of the IGBT 100 by a simple method, the basic characteristics of the device are adversely affected.

On the other hand, according to the IGBT 101 of the present embodiment, latch-up can be prevented without depending on the adjustment of the impurity concentration of the n ⁺ source length SL, the p base length BL, or the p base layer 14. That is, it is possible to prevent the occurrence of latch-up while avoiding the above-described adverse effects related to the basic characteristics of the device.

  Further, by applying the microloading effect, trenches 5V having partially different depths can be formed by a single trench etching.

Although the IGBT has been described in the present embodiment, the same effect as that of the present embodiment can be obtained in the MOSFET by using a structure in which the p ⁺ collector layer 6 is not provided in the structure of the IGBT 101 (FIGS. 2 to 4). Obtainable.

(Embodiment 2)
With reference to FIGS. 17-20, the structure of IGBT102 as a power semiconductor device of this Embodiment is demonstrated. FIG. 17 is a view showing the transistor cell of the IGBT 102 from the emitter side, and in FIG. 17, the emitter electrode 11, the interlayer insulating film 10, and the gate insulating film 9 are not shown for easy understanding of the drawing.

The IGBT 102 includes a collector electrode 12 (first electrode), an emitter electrode 11 (second electrode), a gate insulating film 9, a gate electrode ES, an interlayer insulating film 10, and a semiconductor layer. The semiconductor layer is provided on the collector electrode 12, and includes a p-type (second conductivity type) p ⁺ collector layer 6 (fourth layer) and an n-type (first conductivity type) n ⁺ buffer layer 7. An n-type drift layer 8V (first layer), a p-type (second conductivity type) p base layer 14 (second layer), and a third layer. The third layer has an n-type n ⁺ source region 2 (first region) and a p-type p ⁺ contact region 3 (second region).

The drift layer 8V is provided on the n ⁺ buffer layer 7. That is, the drift layer 8V is provided on the collector electrode 12 via the p ⁺ collector layer 6 and the n ⁺ buffer layer 7. The drift layer 8V has a low concentration region 8m (first low concentration region) and a high concentration region 8p (first high concentration region). As shown in FIG. 18, the high concentration region 8p is buried on the emitter electrode 11 side of the low concentration region 8m, and has a higher impurity concentration than the impurity concentration of the low concentration region 8m.

  The thickness of the high concentration region 8p is, for example, 7 μm in the present embodiment. The high concentration region 8p can be formed, for example, by implanting phosphorus using a mask pattern with high energy of MeV level.

The p base layer 14 is provided on the drift layer 8V.
A trench 5S having a width W1 is formed on the surface opposite to the surface facing the collector electrode 12 of the semiconductor layer (the surface shown in FIG. 17). The gate insulating film 9 covers the inner wall of the trench 5S. The gate electrode ES is embedded in the trench 5S via the gate insulating film 9.

Gate electrode ES includes a first portion (portion shown in FIG. 19) that penetrates n ⁺ source region 2 and p base layer 14 and enters drift layer 8V, p ⁺ contact region 3 and p base layer 14. And a second portion (a portion shown in FIG. 20) penetrating into the drift layer 8V. These first and second portions are integrally formed. That is, the gate electrode ES is provided so as to cross the stripe-like arrangement of the n ⁺ source region 2 and the p ⁺ contact region 3 in plan view. Thus, the n ⁺ source region 2 and the p ⁺ contact region 3 are configured to have the same potential.

The first portion of the gate electrode ES (portion shown by FIG. 19) is, via a gate insulating film 9, and not via the high-density region 8 p, covered by the low concentration region 8m of the drift layer 8V ing. A second portion (portion shown in FIG. 20) of the gate electrode ES is covered with the high concentration region 8p of the drift layer 8V via the gate insulating film 9. The high concentration region 8p is covered with the low concentration region 8m.

Emitter electrode 11 is in contact with each of n ⁺ source region 2 and p ⁺ contact region 3 at the position of emitter contact 4 (FIGS. 3 and 4). The emitter electrode 11 is insulated from the gate electrode ES by the interlayer insulating film 10.

Referring to FIGS. 19 and 21, when the peak concentration in the impurity concentration profile along arrow XXI (FIG. 19) is exemplified as the number of ions per unit volume, the peak concentration of n ⁺ source region 2 is 1 × 10 ¹⁹ / cm. ^3. The peak concentration of the p base layer 14 is 5 × 10 ¹⁷ / cm ³ , the peak concentration of the low concentration region 8m of the drift layer 8V is 1.5 × 10 ¹⁴ / cm ³ , and the peak concentration of the n ⁺ buffer layer 7 is 1. peak concentration of ^{× 10 16 / cm 3, p} + collector layer 6 is 1 × 10 ¹⁹ / cm ^3.

Referring to FIGS. 20 and 22, when the peak concentration in the impurity concentration profile along arrow XXII (FIG. 20) is exemplified as the number of ions per unit volume, the peak concentration of p ⁺ contact region 3 is 1 × 10 ¹⁹ / cm. ^3. The peak concentration of the p base layer 14 is 5 × 10 ¹⁷ / cm ³ , the peak concentration of the high concentration region 8p of the drift layer 8V is 1 × 10 ¹⁵ / cm ³ , and the peak concentration of the low concentration region 8m of the drift layer 8V is 1.5 × 10 ¹⁴ / cm ^3, the peak concentration of n ⁺ buffer layer 7 is ^{1 × 10 16 / cm 3,} p + peak concentration of the collector layer 6 is 1 × 10 ¹⁹ / cm ^3.

  Since the configuration other than the above is substantially the same as the configuration of the first embodiment described above, the same or corresponding elements are denoted by the same reference numerals, and description thereof is not repeated.

According to the present embodiment, high concentration region 8p is provided only at the bottom of trench 5S (FIG. 17) that penetrates p ⁺ contact region 3 (the portion shown in FIG. 20). . As a result, the electric field in the strong electric field region 16 (FIG. 23) is smaller than the electric field in the strong electric field region 23 (FIG. 24). For this reason, the hole current at the time of turn-off mainly passes the path indicated by the arrow 20 among the paths indicated by the arrow 17 (FIG. 23) and the arrow 20 (FIG. 24). That is, the hole current passing through the path indicated by the arrow 17 becomes small due to the difference in electric field strength. As a result, the same effect as in the first embodiment can be obtained. That is, the latch-up of the IGBT 102 is suppressed.

Although the IGBT has been described in the present embodiment, the same effect as in the present embodiment can be obtained in the MOSFET by using a structure in which the p ⁺ collector layer 6 is not provided in the structure of the IGBT 102.

(Embodiment 3)
With reference to FIGS. 25 to 28, the configuration of IGBT 103 as the power semiconductor device of the present embodiment will be described. FIG. 25 is a view showing the transistor cell of the IGBT 103 from the emitter side. In FIG. 25, the emitter electrode 11, the interlayer insulating film 10, and the gate insulating film 9 are not shown for easy viewing.

The IGBT 103 includes a collector electrode 12 (first electrode), an emitter electrode 11 (second electrode), a gate insulating film 9, a gate electrode EV, an interlayer insulating film 10, and a semiconductor layer. The semiconductor layer is provided on the collector electrode 12, and includes a p-type (second conductivity type) p ⁺ collector layer 6 (fourth layer) and an n-type (first conductivity type) n ⁺ buffer layer 7. An n-type drift layer 8V (first layer), a p-type (second conductivity type) p base layer 14 (second layer), and a third layer. The third layer has an n-type n ⁺ source region 2 (first region) and a p-type p ⁺ contact region 3 (second region).

The first portion of the gate electrode EV 1 (portion shown by FIG. 27), a gate insulating film 9, and not via the high-density region 8 p, covered by the low concentration region 8m of the drift layer 8V ing. The second portion 13 (portion shown in FIG. 28) of the gate electrode EV is covered with the high concentration region 8p (first high concentration region) of the drift layer 8V via the gate insulating film 9. . The high concentration region 8p is covered with a low concentration region 8m (first low concentration region).

  In the present embodiment, the thickness of the high concentration region 8p is, for example, 8 μm. The high concentration region 8p can be formed, for example, by implanting phosphorus using a mask pattern with high energy of MeV level.

  Since the configuration other than the above is substantially the same as the configuration of the first or second embodiment described above, the same or corresponding elements are denoted by the same reference numerals, and description thereof will not be repeated.

According to the present embodiment, the same effect as in the first and second embodiments can be obtained.
(Embodiment 4)
With reference to FIGS. 29 to 34, the configuration of IGBT 104 as the power semiconductor device of the present embodiment will be described. FIG. 29 is a diagram showing the transistor cell of the IGBT 104 from the emitter side. In FIG. 29 , the emitter electrode 11, the interlayer insulating film 10, and the gate insulating film 9 are not shown for easy understanding of the drawing.

The IGBT 104 includes a collector electrode 12 (first electrode), an emitter electrode 11 (second electrode), a gate insulating film 9, a gate electrode ES, an interlayer insulating film 10, and a semiconductor layer. The semiconductor layer is provided on the collector electrode 12, and includes a p-type (second conductivity type) p ⁺ collector layer 6 (fourth layer) and an n-type (first conductivity type) n ⁺ buffer layer 7. A drift layer 8W (first layer), a p-type (second conductivity type) p base layer 14 (second layer), and a third layer. The third layer has an n-type n ⁺ source region 2 (first region) and a p-type p ⁺ contact region 3 (second region).

Drift layer 8W is provided on n ⁺ buffer layer 7. That is, drift layer 8W is provided on collector electrode 12 with p ⁺ collector layer 6 and n ⁺ buffer layer 7 interposed therebetween. The drift layer 8W includes an n-type low concentration region 8m (first low concentration region), an n-type high concentration region 8p (first high concentration region), and a p-type low concentration region 25. . As shown in FIG. 32, the high concentration region 8p is buried on the emitter electrode 11 side of the low concentration region 8m, and has a higher impurity concentration than the impurity concentration of the low concentration region 8m.

  As shown in FIG. 31, the low concentration region 25 is disposed on the emitter electrode 11 side of the low concentration region 8 m and has a lower impurity concentration than the impurity concentration of the p base layer 14. The low concentration region 25 can be formed, for example, by implanting boron using a mask pattern with high energy of MeV level.

The p base layer 14 is provided on the drift layer 8W.
Gate electrode ES includes a first portion (portion shown in FIG. 31) that penetrates n ⁺ source region 2 and p base layer 14 and enters drift layer 8W, p ⁺ contact region 3 and p base layer 14. And a second portion (portion shown in FIG. 32) penetrating into the drift layer 8W. These first and second portions are integrally formed. That is, the gate electrode ES is provided so as to cross the stripe-like arrangement of the n ⁺ source region 2 and the p ⁺ contact region 3 in plan view. Thus, the n ⁺ source region 2 and the p ⁺ contact region 3 are configured to have the same potential.

  A second portion (portion shown in FIG. 32) of the gate electrode ES is covered with the high concentration region 8p of the drift layer 8W via the gate insulating film 9. The high concentration region 8p is covered with the low concentration region 8m. A first portion (portion shown in FIG. 31) of the gate electrode ES is covered with the low concentration region 25 of the drift layer 8W via the gate insulating film 9.

Referring to FIG. 31 and FIG. 33, when the peak concentration in the impurity concentration profile along arrow XXXIII (FIG. 31) is exemplified as the number of ions per unit volume, the peak concentration of n ⁺ source region 2 is 1 × 10 ¹⁹ / cm. ^3. The peak concentration of the p base layer 14 is 5 × 10 ¹⁷ / cm ³ , the peak concentration of the low concentration region 25 of the drift layer 8W is 3 × 10 ¹⁴ / cm ³ , and the peak concentration of the low concentration region 8m of the drift layer 8W is 1.5 × 10 ¹⁴ / cm ^3, the peak concentration of n ⁺ buffer layer 7 is ^{1 × 10 16 / cm 3,} p + peak concentration of the collector layer 6 is 1 × 10 ¹⁹ / cm ^3.

Referring to FIGS. 32 and 34, when the peak concentration in the impurity concentration profile along arrow XXXIV (FIG. 32) is exemplified as the number of ions per unit volume, the peak concentration of p ⁺ contact region 3 is 1 × 10 ¹⁹ / cm. ^3. The peak concentration of the p base layer 14 is 5 × 10 ¹⁷ / cm ³ , the peak concentration of the high concentration region 8p of the drift layer 8W is 1 × 10 ¹⁵ / cm ³ , and the peak concentration of the low concentration region 8m of the drift layer 8W is 1.5 × 10 ¹⁴ / cm ^3, the peak concentration of n ⁺ buffer layer 7 is ^{1 × 10 16 / cm 3,} p + peak concentration of the collector layer 6 is 1 × 10 ¹⁹ / cm ^3.

  Since the configuration other than the above is substantially the same as the configuration of the second embodiment described above, the same or corresponding elements are denoted by the same reference numerals, and description thereof will not be repeated.

  According to the present embodiment, the same effect as in the second embodiment can be obtained. Further, the first portion of the gate electrode ES (portion shown in FIG. 31) is covered with the low concentration region 25, so that the channel length is extended. For this reason, the electric field at the bottom of the trench during turn-off can be relaxed without increasing the on-voltage.

Although the IGBT has been described in the present embodiment, the same effect as in the present embodiment can be obtained in the MOSFET by using a structure in which the p ⁺ collector layer 6 is not provided in the structure of the IGBT 104.

(Embodiment 5)
With reference to FIGS. 35 to 38, the configuration of IGBT 105 as the power semiconductor device of the present embodiment will be described. FIG. 35 is a view showing the transistor cell of the IGBT 105 from the emitter side, and in FIG. 35, the emitter electrode 11, the interlayer insulating film 10, and the gate insulating film 9 are not shown for easy understanding of the drawing.

The IGBT 105 includes a collector electrode 12 (first electrode), an emitter electrode 11 (second electrode), a gate insulating film 9, a gate electrode ES, an interlayer insulating film 10, and a semiconductor layer. The semiconductor layer is provided on the collector electrode 12, and includes a p-type (second conductivity type) p ⁺ collector layer 6 (fourth layer) and an n-type (first conductivity type) n ⁺ buffer layer 7. And an n-type drift layer 8 (first layer), a p-type (second conductivity type) p base layer 14V (second layer), and a third layer. The third layer has an n-type n ⁺ source region 2 (first region) and a p-type p ⁺ contact region 3 (second region).

  The p base layer 14V is provided on the drift layer 8, and has a normal concentration base region 14n (normal concentration region) and a high concentration base region 14p (second high concentration region). The high concentration base region 14p has a higher impurity concentration than the impurity concentration of the normal concentration base region 14n.

Each of the normal concentration base region 14n and the high concentration base region 14p is arranged immediately below the n ⁺ source region 2 and the p ⁺ contact region 3 in the semiconductor layer, as shown in FIG. That is, the normal concentration base region 14n and the high concentration base region 14p are formed in a stripe shape, similar to the n ⁺ source region 2 and the p ⁺ contact region 3.

More specifically, each of the normal concentration base region 14n and the high concentration base region 14p has a planar shape similar to that of the n ⁺ source region 2 and the p ⁺ contact region 3 on the emitter side. The p base layer 14V has such a shape that the normal concentration base region 14n gradually erodes into the high concentration base region 14p from the emitter side to the collector side. Such a high-concentration base region 14p can be formed, for example, by implanting boron with energy of 150 keV using a mask pattern having an opening 26 (FIG. 35).

  A trench 5S having a width W1 is formed on the surface opposite to the surface facing the collector electrode 12 of the semiconductor layer (the surface shown in FIG. 35). The gate insulating film 9 covers the inner wall of the trench 5S. The gate electrode ES is embedded in the trench 5S via the gate insulating film 9.

The gate electrode ES includes a first portion (portion shown in FIG. 37) penetrating the n ⁺ source region 2 and the normal concentration base region 14n and entering the drift layer 8, and the p ⁺ contact region 3 and the high concentration base. And a second portion (portion shown in FIG. 38) that penetrates the region 14p and enters the drift layer 8. These first and second portions are integrally formed. That is, the gate electrode ES is provided so as to cross the stripe-like arrangement of the n ⁺ source region 2 and the p ⁺ contact region 3 in plan view. Thus, the n ⁺ source region 2 and the p ⁺ contact region 3 are configured to have the same potential.

  The first portion (portion shown in FIG. 37) of the gate electrode ES is covered with the normal concentration base region 14n via the gate insulating film 9. The second portion (portion shown in FIG. 38) of the gate electrode ES is covered with the high concentration base region 14p via the gate insulating film 9.

Referring to FIGS. 37 and 39, when the peak concentration in the impurity concentration profile along arrow XXXIX (FIG. 37) is exemplified as the number of ions per unit volume, the peak concentration of n ⁺ source region 2 is 1 × 10 ¹⁹ / cm. ^3. The peak concentration of the normal concentration base region 14n of the p base layer 14V is 5 × 10 ¹⁷ / cm ³ , the peak concentration of the drift layer 8 is 1.5 × 10 ¹⁴ / cm ³ , and the peak concentration of the n ⁺ buffer layer 7 is The peak concentration of 1 × 10 ¹⁶ / cm ³ and the p ⁺ collector layer 6 is 1 × 10 ¹⁹ / cm ³ .

Referring to FIGS. 38 and 40, when the peak concentration in the impurity concentration profile along arrow XL (FIG. 38) is exemplified as the number of ions per unit volume, the peak concentration of p ⁺ contact region 3 is 1 × 10 ¹⁹ / cm. ^3. The peak concentration of the high concentration base region 14p of the p base layer 14V is 1 × 10 ¹⁸ / cm ³ , the peak concentration of the drift layer 8 is 1.5 × 10 ¹⁴ / cm ³ , and the peak concentration of the n ⁺ buffer layer 7 is The peak concentration of 1 × 10 ¹⁶ / cm ³ and the p ⁺ collector layer 6 is 1 × 10 ¹⁹ / cm ³ .

  Since the configuration other than the above is substantially the same as the configuration of the first embodiment described above, the same or corresponding elements are denoted by the same reference numerals, and description thereof is not repeated.

  According to the present embodiment, the high-concentration base region 14p (FIG. 42) in the p base layer 14V has a low p base resistance, so that the hole current easily flows to the emitter side as indicated by the arrow 29. . As a result, the hole current at the time of turn-off flows more through the path indicated by the arrow 29 (FIG. 42) than the path indicated by the arrow 28 (FIG. 41). Therefore, the current supplied to the base of the parasitic npn transistor 120 (FIG. 5) is reduced, and the latch-up of the IGBT 105 is suppressed.

  Further, in the p base layer 14V, the impurity concentration of the portion of the normal concentration base region 14n (FIG. 41) can be made the same as the impurity concentration of the p base layer 14 (FIG. 15) in the IGBT 100 of the comparative example. As a result, the electrical characteristics (threshold voltage, etc.) of the IGBT 105 can be made the same as those of the IGBT 100 of the comparative example.

Although the IGBT has been described in the present embodiment, the same effect as in the present embodiment can be obtained in the MOSFET by using a structure in which the p ⁺ collector layer 6 is not provided in the structure of the IGBT 105.

(Embodiment 6)
With reference to FIGS. 43 to 46, the configuration of IGBT 106 as the power semiconductor device of the present embodiment will be described. FIG. 43 is a diagram showing the transistor cell of the IGBT 106 from the emitter side. In FIG. 43, the emitter electrode 11, the interlayer insulating film 10, and the gate insulating film 9 are not shown for easy viewing.

The IGBT 106 includes a collector electrode 12 (first electrode), an emitter electrode 11 (second electrode), a gate insulating film 9, a gate electrode EV, an interlayer insulating film 10, and a semiconductor layer. The semiconductor layer is provided on the collector electrode 12, and includes a p-type (second conductivity type) p ⁺ collector layer 6 (fourth layer) and an n-type (first conductivity type) n ⁺ buffer layer 7. And an n-type drift layer 8 (first layer), a p-type (second conductivity type) p base layer 14V (second layer), and a third layer. The third layer has an n-type n ⁺ source region 2 (first region) and a p-type p ⁺ contact region 3 (second region).

  The first portion 1 (portion shown in FIG. 45) of the gate electrode EV is covered with the normal concentration base region 14n through the gate insulating film 9. Further, the second portion 13 (portion shown in FIG. 46) of the gate electrode EV is covered with the high-concentration base region 14p through the gate insulating film 9.

  Since the configuration other than the above is substantially the same as the configuration of the first or fifth embodiment described above, the same or corresponding elements are denoted by the same reference numerals, and description thereof will not be repeated.

  According to the present embodiment, in addition to the same effects as in the fifth embodiment, the same effects as in the first embodiment can be obtained. Therefore, the latch-up of the IGBT 106 is further suppressed.

(Embodiment 7)
With reference to FIGS. 47 to 50, the configuration of IGBT 107 as the power semiconductor device of the present embodiment will be described. FIG. 47 is a view showing the transistor cell of the IGBT 107 from the emitter side. In FIG. 47, the emitter electrode 11, the interlayer insulating film 10, and the gate insulating film 9 are not shown in order to make the drawing easier to see.

The IGBT 107 includes a collector electrode 12 (first electrode), an emitter electrode 11 (second electrode), a gate insulating film 9, a gate electrode ES, an interlayer insulating film 10, and a semiconductor layer. The semiconductor layer is provided on the collector electrode 12, and includes a p-type (second conductivity type) p ⁺ collector layer 6 (fourth layer) and an n-type (first conductivity type) n ⁺ buffer layer 7. An n-type drift layer 8 (first layer), a p-type (second conductivity type) p base layer 14W (second layer), and a third layer. The third layer has an n-type n ⁺ source region 2 (first region) and a p-type p ⁺ contact region 3 (second region).

The p base layer 14W is provided on the drift layer 8, and has a base region 14b and a retrograde region 14q. Retrograde region 14q is provided in a region where p ⁺ contact region 3 is located in plan view (FIG. 47). The retrograde region 14q is formed in a retrograde structure in the base region 14b from the emitter electrode 11 side (the upper side in FIG. 48) of the semiconductor layer. Details of the retrograde structure will be described later.

  The retrograde region 14q can be formed, for example, by implanting boron using a mask pattern having an opening 26 (FIG. 47) at a high energy of MeV level.

Gate electrode ES includes a first portion (portion shown in FIG. 49) penetrating n ⁺ source region 2 and base region 14b of p base layer 14W and entering drift layer 8, and p ⁺ contact regions 3 and p. And a second portion (portion shown in FIG. 50) penetrating through the base layer 14W and entering the drift layer 8. These first and second portions are integrally formed. That is, the gate electrode ES is provided so as to cross the stripe-like arrangement of the n ⁺ source region 2 and the p ⁺ contact region 3 in plan view. Thus, the n ⁺ source region 2 and the p ⁺ contact region 3 are configured to have the same potential.

Referring to FIGS. 49 and 51, when the peak concentration in the impurity concentration profile along arrow LI (FIG. 49) is exemplified as the number of ions per unit volume, the peak concentration of n ⁺ source region 2 is 1 × 10 ¹⁹ / cm. ^3. The peak concentration of the base region 14b of the p base layer 14W is 5 × 10 ¹⁷ / cm ³ , the peak concentration of the drift layer 8 is 1.5 × 10 ¹⁴ / cm ³ , and the peak concentration of the n ⁺ buffer layer 7 is 1 ×. The peak concentration of 10 ¹⁶ / cm ³ and the p ⁺ collector layer 6 is 1 × 10 ¹⁹ / cm ³ .

Referring to FIGS. 50 and 52, when the peak concentration in the impurity concentration profile along arrow LII (FIG. 50) is exemplified as the number of ions per unit volume, the peak concentration of p ⁺ contact region 3 is 1 × 10 ¹⁹ / cm. ^3, p base layer 14 peak concentration of the base region 14b of W is ^{5 × 10 17 / cm 3,} p base layer the peak concentration of retrograde region 14q of 14W is 1 × 10 ¹⁸ / cm ^3, the peak concentration of the drift layer 8 Is 1.5 × 10 ¹⁴ / cm ³ , the peak concentration of the n ⁺ buffer layer 7 is 1 × 10 ¹⁶ / cm ³ , and the peak concentration of the p ⁺ collector layer 6 is 1 × 10 ¹⁹ / cm ³ . As indicated by the arrows in FIG. 52, the p base layer 14W has an impurity concentration profile, that is, a retrograde profile in which the impurity concentration increases as Z increases, that is, as the depth increases. In other words, the p base layer 14 has a retrograde structure.

  Since the configuration other than the above is substantially the same as the configuration of the first embodiment described above, the same or corresponding elements are denoted by the same reference numerals, and description thereof is not repeated.

According to the present embodiment, the retrograde region 14q is provided in the region immediately below p ⁺ contact region 3 in FIG. 50, so that the effect of preventing latch-up can be obtained as in the fifth embodiment (FIG. 38). It is done. In addition, since the retrograde structure can particularly reduce the p base resistance of the p base layer 14W in the vicinity of the n ⁻ drift layer 8, this effect can be further enhanced.

Although the IGBT has been described in the present embodiment, the same effect as in the present embodiment can be obtained in the MOSFET by using a structure in which the p ⁺ collector layer 6 is not provided in the structure of the IGBT 107.

(Embodiment 8)
Referring mainly to FIG. 53, IGBT 108 as the power semiconductor device of the present embodiment has high-concentration base region 14d instead of high-concentration base region 14p (FIG. 38) of the fifth embodiment. . Unlike the high concentration base region 14p, the high concentration base region 14d is formed deeper than the gate electrode ES.

  The planar pattern of the high concentration base region 14d is the same as the planar pattern of the high concentration base region 14p.

  Since the configuration other than the above is substantially the same as the configuration of the fifth embodiment described above, the same or corresponding elements are denoted by the same reference numerals, and description thereof is not repeated.

According to the present embodiment, latch-up can be prevented as in the fifth embodiment. However, the main breakdown voltage decreases as the effective thickness Te (FIG. 53) of the n ⁻ drift layer 8 decreases. In order to prevent the main breakdown voltage from decreasing, it is preferable to use a high-concentration base region 14p formed shallower than the gate electrode ES as shown in FIG.

(Embodiment 9)
With reference to FIGS. 54 to 58, the configuration of IGBT 109 as the power semiconductor device of the present embodiment will be described. 54 is a diagram showing the transistor cell of the IGBT 109 from the emitter side. In FIG. 54, the emitter electrode 11, the interlayer insulating film 10, and the gate insulating film 9 are not shown in order to make the drawing easier to see.

The IGBT 109 includes a collector electrode 12 (first electrode), an emitter electrode 11 (second electrode), a gate insulating film 9, a gate electrode ES, an interlayer insulating film 10, and a semiconductor layer. The semiconductor layer is provided on the collector electrode 12, and includes a p-type (second conductivity type) p ⁺ collector layer 6 (fourth layer) and an n-type (first conductivity type) n ⁺ buffer layer 7. An n-type drift layer 8V (first layer), a p-type (second conductivity type) p base layer 14V (second layer), and a third layer. The third layer has an n-type n ⁺ source region 2 (first region) and a p-type p ⁺ contact region 3 (second region). The configuration of drift layer 8V is substantially the same as in the second embodiment, and the configuration of p base layer 14V is substantially the same as in the fifth embodiment.

Referring to FIGS. 57 and 59, when the peak concentration in the impurity concentration profile along arrow LIX (FIG. 57) is exemplified as the number of ions per unit volume, the peak concentration of n ⁺ source region 2 is 1 × 10 ¹⁹ / cm. ^3. The normal concentration base region 14n peak concentration of the p base layer 14V is 5 × 10 ¹⁷ / cm ³ , the peak concentration of the low concentration region 8m of the drift layer 8V is 1.5 × 10 ¹⁴ / cm ³ , and the n ⁺ buffer layer 7 Is 1 × 10 ¹⁶ / cm ³ , and the p ⁺ collector layer 6 has a peak concentration of 1 × 10 ¹⁹ / cm ³ .

Referring to FIGS. 58 and 60, when the peak concentration in the impurity concentration profile along arrow LX (FIG. 58) is exemplified as the number of ions per unit volume, the peak concentration of p ⁺ contact region 3 is 1 × 10 ¹⁹ / cm. ^3. The peak concentration of the high concentration base region 14p of the p base layer 14V is 1 × 10 ¹⁸ / cm ³ , the peak concentration of the high concentration region 8p of the drift layer 8V is 1 × 10 ¹⁵ / cm ³ , and the low concentration of the drift layer 8V. The peak concentration of the region 8m is 1.5 × 10 ¹⁴ / cm ³ , the peak concentration of the n ⁺ buffer layer 7 is 1 × 10 ¹⁶ / cm ³ , and the peak concentration of the p ⁺ collector layer 6 is 1 × 10 ¹⁹ / cm ³ . is there.

  Since the configuration other than the above is substantially the same as the configuration of the second or fifth embodiment described above, the same or corresponding elements are denoted by the same reference numerals, and description thereof will not be repeated.

  According to the present embodiment, the effect of preventing latch-up can be obtained as in the second embodiment. Further, since the same effect as in the fifth embodiment can be obtained, latch-up can be prevented more reliably.

(Embodiment 10)
With reference to FIGS. 61 to 65, the configuration of IGBT 110 as the power semiconductor device of the present embodiment will be described. FIG. 61 is a diagram showing the transistor cell of the IGBT 110 from the emitter side. In FIG. 61, the emitter electrode 11, the interlayer insulating film 10, and the gate insulating film 9 are not shown for easy understanding of the drawing.

The IGBT 110 includes a collector electrode 12 (first electrode), an emitter electrode 11 (second electrode), a gate insulating film 9, a gate electrode ES, an interlayer insulating film 10, and a semiconductor layer. The semiconductor layer is provided on the collector electrode 12, and includes a p-type (second conductivity type) p ⁺ collector layer 6 (fourth layer) and an n-type (first conductivity type) n ⁺ buffer layer 7. An n-type drift layer 8V (first layer), a p-type (second conductivity type) p base layer 14W (second layer), and a third layer. The third layer has an n-type n ⁺ source region 2 (first region) and a p-type p ⁺ contact region 3 (second region). The configuration of drift layer 8V is substantially the same as in the second embodiment, and the configuration of p base layer 14W is substantially the same as in the seventh embodiment.

Referring to FIGS. 64 and 66, when the peak concentration in the impurity concentration profile along arrow LXVI (FIG. 64) is exemplified as the number of ions per unit volume, the peak concentration of n ⁺ source region 2 is 1 × 10 ¹⁹ / cm. ^3. The peak concentration of the base region 14b of the p base layer 14W is 5 × 10 ¹⁷ / cm ³ , the peak concentration of the low concentration region 8m of the drift layer 8V is 1.5 × 10 ¹⁴ / cm ³ , and the n ⁺ buffer layer 7 The peak concentration is 1 × 10 ¹⁶ / cm ³ , and the peak concentration of the p ⁺ collector layer 6 is 1 × 10 ¹⁹ / cm ³ .

Referring to FIGS. 65 and 67, when the peak concentration in the impurity concentration profile along arrow LXVII (FIG. 65) is exemplified as the number of ions per unit volume, the peak concentration of p ⁺ contact region 3 is 1 × 10 ¹⁹ / cm. ^3. The peak concentration of the base region 14b of the p base layer 14W is 5 × 10 ¹⁷ / cm ³ , the peak concentration of the retrograde region 14q of the p base layer 14W is 1 × 10 ¹⁸ / cm ³ , and the high concentration region of the drift layer 8V. The peak concentration of 8p is 1 × 10 ¹⁵ / cm ³ , the peak concentration of the low concentration region 8m of the drift layer 8V is 1.5 × 10 ¹⁴ / cm ³ , and the peak concentration of the n ⁺ buffer layer 7 is 1 × 10 ¹⁶ / cm 3. ^3. The peak concentration of the p ⁺ collector layer 6 is 1 × 10 ¹⁹ / cm ³ .

  Since the configuration other than the above is substantially the same as the configuration of the second or seventh embodiment described above, the same or corresponding elements are denoted by the same reference numerals, and description thereof will not be repeated.

  According to the present embodiment, the effect of preventing latch-up can be obtained as in the second embodiment. Further, since the same effect as in the seventh embodiment can be obtained, latch-up can be prevented more reliably.

(Embodiment 11)
With reference to FIGS. 68 to 72, the configuration of IGBT 111 as the power semiconductor device of the present embodiment will be described. 68 is a view showing the transistor cell of the IGBT 111 from the emitter side. In FIG. 68, the emitter electrode 11, the interlayer insulating film 10, and the gate insulating film 9 are not shown in order to make the drawing easier to see.

The IGBT 111 includes a collector electrode 12 (first electrode), an emitter electrode 11 (second electrode), a gate insulating film 9, a gate electrode EV, an interlayer insulating film 10, and a semiconductor layer. The semiconductor layer is provided on the collector electrode 12, and includes a p-type (second conductivity type) p ⁺ collector layer 6 (fourth layer) and an n-type (first conductivity type) n ⁺ buffer layer 7. An n-type drift layer 8V (first layer), a p-type (second conductivity type) p base layer 14V (second layer), and a third layer. The third layer has an n-type n ⁺ source region 2 (first region) and a p-type p ⁺ contact region 3 (second region).

  The configurations of gate electrode EV, drift layer 8V, and p base layer 14V are substantially the same as those in the first, second, and fifth embodiments.

Referring to FIGS. 71 and 73, when the peak concentration in the impurity concentration profile along arrow LXXIII (FIG. 71) is exemplified as the number of ions per unit volume, the peak concentration of n ⁺ source region 2 is 1 × 10 ¹⁹ / cm. ^3. The peak concentration of the normal concentration base region 14n of the p base layer 14V is 5 × 10 ¹⁷ / cm ³ , the peak concentration of the low concentration region 8m of the drift layer 8V is 1.5 × 10 ¹⁴ / cm ³ , and the n ⁺ buffer layer 7 has a peak concentration of 1 × 10 ¹⁶ / cm ³ and the p ⁺ collector layer 6 has a peak concentration of 1 × 10 ¹⁹ / cm ³ .

Referring to FIGS. 72 and 74, when the peak concentration in the impurity concentration profile along arrow LXXIV (FIG. 72) is exemplified as the number of ions per unit volume, the peak concentration of p ⁺ contact region 3 is 1 × 10 ¹⁹ / cm. ^3. The peak concentration of the high concentration base region 14p of the p base layer 14V is 1 × 10 ¹⁸ / cm ³ , the peak concentration of the high concentration region 8p of the drift layer 8V is 1 × 10 ¹⁵ / cm ³ , and the low concentration of the drift layer 8V. The peak concentration of the region 8m is 1.5 × 10 ¹⁴ / cm ³ , the peak concentration of the n ⁺ buffer layer 7 is 1 × 10 ¹⁶ / cm ³ , and the peak concentration of the p ⁺ collector layer 6 is 1 × 10 ¹⁹ / cm ³ . is there.

  Since the configuration other than the above is substantially the same as the configuration of the first, second, or fifth embodiment described above, the same or corresponding elements are denoted by the same reference numerals, and description thereof will not be repeated.

  According to the present embodiment, the effect of preventing latch-up can be obtained as in the fifth embodiment. In addition, since the same effect as in the first and second embodiments can be obtained, latch-up can be prevented more reliably.

  In the description of each of the above embodiments, the first conductivity type is n-type and the second conductivity type is p-type, but the first conductivity type is p-type and the second conductivity type is n-type. Good.

  In order to obtain a semiconductor layer, for example, a wafer by an epitaxial growth method or an FZ (Floating Zone) method can be used.

  Further, the power semiconductor device is not limited to the IGBT or the MOSFET but may be, for example, a CSTBT.

  The power semiconductor device is, for example, a silicon device, but is not limited thereto. For example, the power semiconductor device may be a silicon carbide device that has been developed recently and is expected to have high efficiency.

  Each embodiment disclosed this time must be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  The present invention can be applied particularly advantageously to a power semiconductor device having a gate electrode embedded in a trench.

ES, EV gate electrode, 1st portion, 2 n ⁺ source region (first region), 3 p ⁺ contact region (second region), 4 emitter contact, 5S, 5V trench, 6 p ⁺ collector layer (Fourth layer), 7 n ⁺ buffer layer, 8, 8V, 8W drift layer (first layer), 8m low concentration region (first low concentration region), 8p high concentration region (first high concentration) Region), 9 gate insulating film, 10 interlayer insulating film, 11 emitter electrode (second electrode), 12 collector electrode (first electrode), 13 second part, 14, 14V, 14W p base layer (second 14n, normal concentration base region (normal concentration region), 14d, 14p high concentration base region (second high concentration region), 101-111 IGBT (power semiconductor device).

Claims (13)

A first electrode;
A semiconductor layer provided on the first electrode,
A trench is formed on a surface of the semiconductor layer opposite to the surface facing the first electrode;
The semiconductor layer is
A first conductivity type first layer provided on the first electrode;
A second layer of a second conductivity type provided on the first layer;
And a third layer provided on the second layer and having a first region of the first conductivity type and a second region of the second conductivity type, and the first and second layers A second electrode in contact with each of the regions;
A gate insulating film covering the inner wall of the trench;
A gate electrode embedded in the trench through the gate insulating film,
The gate electrode penetrates the first region and the second layer and penetrates the first layer, and penetrates the second region and the second layer. A second portion penetrating the first layer,
The power semiconductor device, wherein the second portion penetrates deeply into the first layer as compared to a depth at which the first portion penetrates into the first layer.   2. The power use according to claim 1, wherein a portion of the trench in which the second portion of the gate electrode is embedded is wider than a portion of the trench in which the first portion of the gate electrode is embedded. Semiconductor device.   3. The power semiconductor device according to claim 1, wherein the semiconductor layer includes a fourth layer of the second conductivity type between the first electrode and the first layer. A first electrode;
A semiconductor layer provided on the first electrode,
A trench is formed on a surface of the semiconductor layer opposite to the surface facing the first electrode;
The semiconductor layer is
A first conductivity type first layer provided on the first electrode;
A second layer of a second conductivity type provided on the first layer;
And a third layer provided on the second layer and having a first region of the first conductivity type and a second region of the second conductivity type, and the first and second layers A second electrode in contact with each of the regions;
A gate insulating film covering the inner wall of the trench;
A gate electrode embedded in the trench through the gate insulating film,
The gate electrode penetrates the first region and the second layer and penetrates the first layer, and penetrates the second region and the second layer. A second portion penetrating the first layer,
The first layer is
A first low concentration region;
Only the second portion of the first and second portions of the gate electrode is covered via the gate insulating film, and has a higher impurity concentration than the impurity concentration of the first low concentration region. A power semiconductor device including a first high concentration region.   The power semiconductor device according to claim 4, wherein the semiconductor layer includes a fourth layer of the second conductivity type between the first electrode and the first layer.   The power semiconductor according to claim 4 or 5, wherein the second portion penetrates deeply into the first layer as compared to a depth at which the first portion penetrates into the first layer. apparatus. The second layer is
The base region,
Provided only in the region where the second region is located among the first and second regions of the third layer in plan view, and from the second electrode side of the semiconductor layer into the base region The power semiconductor device according to claim 4, further comprising a retrograde region formed of a retrograde structure. The second layer is
A normal concentration region covering the first portion via the gate insulating film;
6. The device according to claim 4, further comprising: a second high concentration region that covers the second portion via the gate insulating film and has a higher impurity concentration than the impurity concentration of the normal concentration region. Power semiconductor device.   The power semiconductor device according to claim 8, wherein the second portion penetrates deeply into the first layer as compared to a depth at which the first portion penetrates into the first layer. A first electrode;
A semiconductor layer provided on the first electrode,
A trench is formed on a surface of the semiconductor layer opposite to the surface facing the first electrode;
The semiconductor layer is
A first layer provided on the first electrode;
A second layer of a second conductivity type provided on the first layer;
And a third layer provided on the second layer and having a first region of the first conductivity type and a second region of the second conductivity type, and the first and second layers A second electrode in contact with each of the regions;
A gate insulating film covering the inner wall of the trench;
A gate electrode embedded in the trench through the gate insulating film,
The gate electrode penetrates the first region and the second layer and penetrates the first layer, and penetrates the second region and the second layer. A second portion penetrating the first layer,
The first layer is
A first low concentration region of the first conductivity type;
A first high-concentration region of the first conductivity type that covers the second portion through the gate insulating film and has a higher impurity concentration than the impurity concentration of the first low-concentration region;
A second low concentration region of the second conductivity type that covers the first portion through the gate insulating film and has a lower impurity concentration than the impurity concentration of the second layer. Semiconductor device.   The power semiconductor device according to claim 10, wherein the semiconductor layer includes a fourth layer of the second conductivity type between the first electrode and the first layer. A first electrode;
A semiconductor layer provided on the first electrode,
A trench is formed on a surface of the semiconductor layer opposite to the surface facing the first electrode;
The semiconductor layer is
A first conductivity type first layer provided on the first electrode;
A second layer of a second conductivity type provided on the first layer;
And a third layer provided on the second layer and having a first region of the first conductivity type and a second region of the second conductivity type, and the first and second layers A second electrode in contact with each of the regions;
A gate insulating film covering the inner wall of the trench;
A gate electrode embedded in the trench through the gate insulating film,
The gate electrode penetrates the first region and the second layer and penetrates the first layer, and penetrates the second region and the second layer. A second portion penetrating the first layer,
The second layer is
A normal concentration region covering the first portion via the gate insulating film;
A second high-concentration region that covers the second part through the gate insulating film and has a higher impurity concentration than the impurity concentration of the normal concentration region,
The power semiconductor device, wherein the second portion penetrates deeply into the first layer as compared to a depth at which the first portion penetrates into the first layer. The power semiconductor device according to claim 12, wherein the semiconductor layer includes a fourth layer of the second conductivity type between the first electrode and the first layer.

Images